This proposal describes research aimed at understanding the immunologic and pathogenic processes involved with the human respiratory pathogen, Legionella pneumophila. Eventually, it is hoped that a vaccine for Legionnaires' disease will emerge from this research effort. Although the scientific literature contains preliminary data on the antigenic structure of this bacterium, detailed study of individual protein antigens at the molecular level and characterization of the role of these antigens in infection has not yet been accomplished. To achieve the overall objective of characterizing the immunobiology of L. pneumophila, the proposal is subdivided into three major aims: 1) identification and characterization of the bacterial protein antigens, 2) characterization of the antibodies that are directed against these antigens, and 3) purification and evaluation of candidate protein vaccines. Antigen characterization will be accomplished by utilizing the newer techniques of radioimmunoprecipitation and immunoblotting to determine which proteins are recognized by sera from convalescing or hyperimmunized experimental animals and convalescing humans with legionnaires' disease. Various serogroups will be studied, with an emphasis on serogroup 1, to search for shared antigens. To complement this approach, Legionella antigens cloned in E. coli will be studied as well. Antibody characterization will be accomplished by raising monospecific and monoclonal antibodies to the major antigens and assessing the role of these antibodies in disease protection as determined by experimental infection models. Strategies for vaccine development will follow results from the experimental infection systems. Protective antibodies will be used to define vaccine candidates; these proteins will be purified by biochemical/immunoaffinity techniques and employed in protection studies.